A simplified method for estimating CO 2 emissions from deforestation is the calculation of carbon stock change by monitoring forest land and periodically summing up the land area and its averaged carbon stock for important forest types. As a feasibility study for applying this methodology to a tropical dry-land forest, we estimated carbon stock and its chronosequential change in 4 carbon pools (aboveground and belowground biomass, deadwood, and litter) of tropical dry-land natural forests in Cambodia. Carbon stock differed among forest types. Most of the carbon stock (84 ± 12% (SD)) existed in tree biomass. Growth of carbon stock has a positive relationship to the carbon stock itself. By moderately classifying forest types, determining averaged tree biomass of each forest type, and using land-area data on each forest type, a reasonably accurate estimation of carbon stock can be expected. However, considering that rapidly progressing deforestation and wood extraction may reduce the carbon stock in forests, systematic sampling with a sufficient number of extra plots and frequent updating of forest land area and averaged carbon stock data are vital for an accurate estimation of CO 2 emissions from forests under pressure of land-use change and forestry activities.Discipline: Forestry and forest products Additional key words: biomass, deforestation, forest degradation, REDD, tropical forest This paper reports the results obtained in the collaborative research project on the "Joint implementation of carbon stock estimation by forest measurement to contribute to sustainable forest management in Cambodia"

No abstract

BackgroundData availability in developing countries is known to be extremely varied and is one of the constraints for setting the national reference levels (RLs) for the REDD-plus (i.e. 'Policy approaches and positive incentives on issues relating to reducing emissions from deforestation and forest degradation in developing countries; and the role of conservation, sustainable management of forests and enhancement of forest carbon stocks in developing countries') under the UNFCCC. Taking Thailand as a case study country, this paper compares three types of RLs, which require different levels of datasets, including a simple historic RL, a projected forest-trend RL, and a business-as-usual (BAU) RL.ResultsOther than the finding that different RLs yielded different estimations on future deforestation areas, the analysis also identified the characteristics of each RL. The historical RL demanded simple data, but can be varied in accordance with a reference year or period. The forest-trend RL can be more reliable than the historical RL, if the country's deforestation trend curve is formed smoothly. The complicated BAU RL is useful as it can demonstrate the additionality of REDD-plus activities and distinguish the country's unintentional efforts.ConclusionsWith the REDD-plus that involves widespread participation, there should be steps from which countries choose the appropriate RL; ranging from simpler to more complex measures, in accordance with data availability in each country. Once registered with REDD-plus, the countries with weak capacity and capability should be supported to enhance the data collection system in that country.

No abstract